METHODOLOGY FOR ESTIMATION OF AUTOMATIC SPEECH RECOGNITION SYSTEM PERFORMANCE
Saint Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences, Saint Petersburg, 199178, Russia; ITMO University, Saint Petersburg, 197101, Russian Federation; Laboratory head; professor
I. S. Kipyatkova
St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences (SPIIRAS), Saint Petersburg, 199178, Russian Federation; senior researcher
Abstract. The state-of-the-art methodology of quantitative evaluation of automatic systems for speech recognition and speaker diarization is presented. Various measures and methods for estimation of automatic speech recognition results by criteria of speech recognition accuracy and signal processing speed are described.
Keywords: automatic speech recognition, speech recognition accuracy, evaluation criteria and measures.